How an India-Pakistan nuclear war could start—and have global consequencesAlan Robock, Owen B. Toon, Charles G. Bardeen, Lili Xia, Hans M. Kristensen, Matthew McKinzie, R. J. Peterson,Cheryl S. Harrison, Nicole S. Lovenduski and Richard P. Turco

ABSTRACTThis article describes how an India-Pakistan nuclear war might come to pass, and what the localand global effects of such a war might be. The direct effects of this nuclear exchange would behorrible; the authors estimate that 50 to 125 million people would die, depending on whether theweapons used had yields of 15, 50, or 100 kilotons. The ramifications for Indian and Pakistanisociety would be major and long lasting, with many major cities largely destroyed and uninhabi-table, millions of injured people needing care, and power, transportation, and financial infrastruc-ture in ruins. But the climatic effects of the smoke produced by an India-Pakistan nuclear war wouldnot be confined to the subcontinent, or even to Asia. Those effects would be enormous and globalin scope.

KEYWORDSNuclear war; South Asia;Kashmir; cold start; tacticalnuclear weapons; nuclearwinter

It is the year 2025, and terrorists attack the IndianParliament. In December 2001, a terrorist attack onthe Indian Parliament resulted in the deaths of 12people, including the 5 terrorists. This time, how-ever, the attacks kill many more members of theIndian government. As happened in January 2002,both sides mobilize and deploy their troops alongthe border between the countries and in the dis-puted area of Kashmir. Because of the high tensionson both sides, skirmishes break out, and there aredeaths on both sides. Since the Indian governmenthas lost so many leaders, the Indian Army decidesto act on its own, crossing the border into Pakistanwith tanks and also the de facto border, known asthe Line of Control, in Kashmir.

Pakistani generals panic and decide that the only waythey can repulse an invasion by the superior Indian forces iswith nuclear weapons. On the first day of the nuclear war,they use 10 tactical atomic bombs – each with a yield of5 kilotons, or less than half the power of the bombdroppedon Hiroshima – inside their own borders, detonating themat low altitude, as air bursts against the Indian tanks. Onthe second day, after Pakistan uses another 15 tacticalnuclear weapons, the Indians figure that if they attackPakistani military targets with nuclear weapons, it mightstop the war. The Indians use 20 strategic weapons deto-nated as airbursts, two over the Pakistani garrison inBahawalpur and 18 above Pakistani airfields and nuclearweapons depots. Unlike Pakistan’s tactical weapons, whichwere used in remote areas, these weapons start immensefires, withmassive smoke emissions that rise into the upper

atmosphere, as happened in Hiroshima after it wasbombed by the United States in 1945, and as happenedin San Francisco in 1906 as the result of fire following anearthquake.

The Indian escalation does not work. Rather thanstopping its nuclear attacks, on the third day Pakistanuses 30 airbursts – 20 above garrisons in Indian cities and10 over Indian naval bases and airfields in urban areas –and launches another 15 tactical nuclear weapons atIndian troops. India responds with nuclear airburstsover 10 Pakistani navy, army, and air force bases, alllocated in urban areas. Now the escalation cannot bestopped. There are anger, panic, miscommunication,and the following of pre-determined protocols on bothsides. Over the next three days, Pakistan uses the rest ofits strategic arsenal, with 120 weapons decimatingIndian cities; India responds with another 70 airbursts,but reserves 100 weapons in its arsenal, thinking thatthey will deter any attack from China and ignoringa tragic reality: The Indian nuclear arsenal had just failedto deter a war with Pakistan that killed tens of millions ofpeople immediately and would create enormous envir-onmental impacts, causing famines that affect millions –or even billions – around the world. Figure 1 shows thelocations of the 250 urban targets in our scenario.

Why an India-Pakistan nuclear war really couldhappen

It is not hard to imagine a skirmish between Indianand Pakistani troops along the Line of Control in

CONTACT Alan Robock robock@envsci.rutgers.eduThis article has been republished with minor changes. These changes do not impact the academic content of the article.

BULLETIN OF THE ATOMIC SCIENTISTS2019, VOL. 75, NO. 6, 273–279https://doi.org/10.1080/00963402.2019.1680049

© 2019 Bulletin of the Atomic Scientists

Kashmir. However, neither country is likely to starta nuclear war because of such a skirmish. In fact,India, unlike Pakistan, has a declared policy of nofirst use of nuclear weapons. Pakistan says it willonly use nuclear weapons if needed to defend itselfshould “conventional” means of warfare fail.However, these countries have fought four conven-tional wars (1947, 1965, 1971, and 1999) and hadmany skirmishes with substantial loss of life sincethe partition of British India in 1947. In early 2019,in fact, following fighting in Kashmir, India invadedPakistan by air, and one of the Indian planes wasdowned inside Pakistan. Fortunately the pilot sur-vived and was returned to India without furtherwarfare. But will we always be so lucky? Just inAugust 2019, the constitutionally guaranteed specialstatus of the state of Jammu and Kashmir wasrepealed by India, and the state was locked downby Indian troops to prevent protest. As of this writ-ing, the situation remains tense, and India mayreorganize the region into two new union territoriesthat will be governed directly by the Indian CentralGovernment rather than their own localgovernments.

To investigate the local and global consequences ofa nuclear war between India, we (Toon et al. 2019)investigated the possible outcomes of such a war byconsidering one specific scenario of how it might start.While it is possible to think of many other story lines, theone we used is plausible. It by no means is intended toplace blame on one side or the other for the initiation orescalation of the conflict, and such an escalation wouldnot result without bad decisions on both sides, whichcould be exacerbated by terrorist attacks within eithercountry, panic, loss of communication, technical failuresin observing systems, hacking, or misinterpretation ofthe actions of the other military.

The scenario described here, we assume, wouldtake place in the year 2025, when each country willpossess about 250 nuclear weapons. In the end,Pakistan will use all its weapons, while India willreserve 100 of them to defend against futureattacks from China, which, after all, is one reasonthe Indians obtained them in the first place.

The direct effects of this nuclear exchange wouldbe horrible; our group (Toon et al. 2019) estimatedthat 50 to 125 million people would die, dependingon whether the weapons used had yields of 15, 50,

Figure 1. Urban targets in our India-Pakistan scenario. Different colors represent different days of the war. No urban targets areattacked on day 1. In dense urban areas, some of the dots overlap, for instance in Karachi on the southern coast of Pakistan. (Figure S1from Toon et al. (2019)).

274 A. ROBOCK ET AL.

or 100 kilotons. (A kiloton is the equivalent of theexplosive power of 1,000 tons of TNT.) The ramifica-tions for Indian and Pakistani society would bemajor and long lasting, with many major cities lar-gely destroyed and uninhabitable, millions ofinjured people needing care, and power, transporta-tion, and financial infrastructure in ruins.

But the climatic effects of the smoke producedby an India-Pakistan nuclear war would not be con-fined to the subcontinent, or even to Asia. Thoseeffects would be enormous and global in scope.

The smoke of war

We calculated the climatic effects of differentamounts of smoke injected into the stratosphereas a result of nuclear war using a state-of-the-artclimate model, as detailed in our study (Toon et al.2019). A nuclear war between the United States andRussia could produce 150 teragrams (one teragramequaling one million tons) of smoke, which wouldcreate nuclear winter, with surface temperaturesbelow freezing even in summer (Coupe et al. 2019).

For the India-Pakistan case, the amount of smokewould depend on how large the strategic weaponsof the two countries might be. We have assumedthat Indian and Pakistani strategic weapons are cur-rently the size of the Hiroshima bomb (approxi-mately 15 kilotons), but by 2025, both countriescould have 50 kiloton or 100 kiloton bombs. Indiatested a weapon with a yield of 40 to 50 kilotons in1998. In the India-Pakistan scenario, we calculated

a total of 16.1 teragrams of black carbon injectedinto the upper atmosphere (11 from India and 5.1from Pakistan) for weapons with yields of 15 kilo-tons; 27.3 teragrams (19.8 from India and 7.5 fromPakistan) for 50 kiloton weapons; and 36.6 teragrams(27.5 from India and 9.1 from Pakistan) for 100 kilo-ton weapons. The smoke would be heated by sun-light and lofted high into the stratosphere, where itcould remain for years, since it doesn’t rain in thestratosphere. Figure 2 shows that global averagetemperature and precipitation would be significantlylowered over the course of years, and Figure 3shows how land and ocean temperatures wouldchange separately, also showing a map of the tem-perature change for the middle scenario (27.3 tera-grams of smoke from 50-kiloton detonations) inthe second year after the war, when there wouldbe the maximum effects.

A nuclear winter would halt agriculture aroundthe world and produce famine for billions of people.Though not of the scale of the US-Russia nuclearwar referenced earlier, all of the three scenariosdescribed in the hypothetical India-Pakistan nuclearwar just described would produce severe effects forperiods of years. We have calculated how food pro-duction would change in China (Xia and Robock2013; Xia et al. 2015) and the United States(Özdoğan et al., 2013) for specific crops for a caseof 5 teragrams of smoke – that is, a case involvingsignificantly less smoke than any of the three India-Pakistan scenarios described here. We are nowusing detailed calculations of how specific food

Prec

ipita

tion

chan

ge (

%)

a b

Tem

pera

ture

cha

nge

(o C)

5 Tg16.1 Tg27.3 Tg36.6 Tg46.8 Tg150Tg

Figure 2. Global average precipitation (a) and global average temperature (b) show the climate response to different amounts of blackcarbon emitted into the upper atmosphere from fires following nuclear war. The vertical purple bar represents the range oftemperatures during the height of the last ice age about 20,000 years ago. (Figure 5 from Toon et al. (2019)).

BULLETIN OF THE ATOMIC SCIENTISTS 275

crops globally would respond to the resulting tem-perature, precipitation, and sunlight reductions forvarious smaller amounts of smoke. Also, ozonewould be destroyed as the rising smoke absorbssunlight and heats the stratosphere (Mills et al.2014), allowing more ultraviolet light to reach theground and creating negative effects that we haveyet to study.

While we wait for agricultural simulations to becompleted, our climate model can calculate a moregeneral measure of environmental health, net pri-mary productivity – a measure of how much carbondioxide is converted to organic plant matterthrough photosynthesis after accounting for plantrespiration. Net primary productivity is thereforea proxy for how much food could be grown on

land and how much food would grow in the oceansfor fish. (See Figure 4.) Based on these results, anyof the India-Pakistan nuclear cases we posit clearlywould cause large reductions in agriculture andfood shortages. Depending on whether peoplehoard food or share, there could be famine formillions or billions of people – even for the smalleramounts of smoke in the scenarios presented here.

Conclusions about limiting and eventuallyeliminating nuclear weapons

We have investigated some of the more well known,as well as some lesser known, horrific consequencesof the hostile use of nuclear weapons. These weap-ons have, in principle, had only one legitimate

a b

c

Figure 3. Decline in global average ocean surface temperature (a) and land surface temperature (b) as a function of time. Color-codingshows the assumed black carbon injections. 1 teragram (Tg) is 1 million tons. Panel C illustrates the global distribution of changes inocean and land surface temperatures averaged over the second calendar year following a conflict beginning in May of year one fora scenario with 50 kt weapons, which results in a 27.3 Tg injection of black carbon. (Figure S6 from Toon et al. (2019)).

276 A. ROBOCK ET AL.

purpose: to deter warfare between nations. Onecould argue that that goal has to date beenachieved, inasmuch as no global military conflictshave occurred since World War II. On the otherhand, the existence of enormous arsenals of nuclearweapons during this time has not prevented terror-ism or countless regional, territorial, and politicallymotivated military actions, taking in aggregatea terrible human toll. It would be foolhardy, ofcourse, to suggest that an effective way to stopwarfare would be to arm all nations with nuclearweapons as local deterrents. Contrarily, we under-stand, in the 21st century, that establishingmechanisms for conflict negotiation and resolutionon a global international basis is the only safe andpractical way to end the carnage. We are notPollyannas. But it should be the mission of everyconcerned citizen, particularly those in positions ofinfluence, to work toward the abolition of nuclear

weapons, within the context of global peace andsecurity mechanisms.

During our lifetimes, we have seen progresstoward this goal, especially through a series of spe-cific nuclear arms treaties among the major nuclearpowers, as well as peace programs and policiesdeveloped by the United Nations. For example, asof this writing, 32 nations have ratified the 2017United Nations Treaty on the Prohibition of NuclearWeapons, while 79 nations have signed the treaty;when 50 nations have ratified it, the treaty will comeinto force. However, the nine current nuclear weap-ons states, and many of their allies, have resisted thiseffort. These privileged countries, in general, want toproceed more slowly and carefully, with steppedreductions in, or stabilization of, existing nucleararsenals. We would certainly applaud a progressiveand well-thought-out nuclear weapons reduction andelimination plan for the world.

a b

c d

Figure 4. Globally-integrated monthly-averaged net primary productivity (NPP) change over the oceans (a) and land masses (b) fordifferent amounts of smoke. NPP is a measure of how much carbon dioxide is converted to organic plant matter throughphotosynthesis after accounting for plant respiration, and is typically expressed as grams of carbon per square meter per year (gC/m2/yr). Panel C gives the global distribution of annual average NPP for the baseline control run. Panel D shows the change from thebaseline averaged over the second calendar year following a nuclear conflict which starts in May of year one for the scenario with 50 ktweapons and a 27 Tg injection of smoke (Figure 6 from Toon et al. (2019)).

BULLETIN OF THE ATOMIC SCIENTISTS 277

But it seems that some nations are instead headed fora replay of the old “Cold War.” Not only has nuclear pro-liferation not ended, but additional countries are consider-ing going nuclear. Instead of extending and expandingexisting treaties, the United States and Russia are choosingto upgrade their arsenals and are talking about new gen-erations of nuclear weaponry more effective than the oldvariety. “Rogue” nations – notably North Korea – are pro-ceeding apace with their nuclear weapons programsdespite hollow claims that they plan to denuclearize. Andcontemporary terrorist groups are seeking nuclear capabil-ity in an increasingly loose global bazaar for such devices.In this situation, and in light of the science we are familiarwith, we must endorse forceful actions to limit and even-tually eliminate nuclear weapons as a means of assuringpeace. There is a way, and it must be achieved.

Acknowledgments

This work is supported by the Open Philanthropy Project.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Alan Robock is a Distinguished Professor in the Department ofEnvironmental Sciences at Rutgers University. He graduatedfrom the University of Wisconsin, Madison, in 1970 with a B.A.in Meteorology, and from the Massachusetts Institute ofTechnology with an S.M. in 1974 and Ph.D. in 1977, both inMeteorology. Before graduate school, he served as a PeaceCorps Volunteer in the Philippines. He is a recipient of theAmerican Meteorological Society Jule G. Charney Medal.Robock was a Lead Author of the 2013 Working Group 1 FifthAssessment Report of the Intergovernmental Panel on ClimateChange (awarded the Nobel Peace Prize in 2007). In 2017 theInternational Campaign to Abolish Nuclear Weapons wasawarded the Nobel Peace Prize for “for its work to draw atten-tion to the catastrophic humanitarian consequences of any useof nuclear weapons and for its groundbreaking efforts toachieve a treaty-based prohibition of such weapons” basedon the work of Robock, Toon, Turco, Xia, and others.

Owen B. Toon is a Professor in the Department of Atmosphericand Oceanic Sciences, and in the Laboratory for Atmosphericand Space Physics at the University of Colorado in Boulder. Heis an American Geophysical Union Roger Revelle Medalist andan American Meteorological Society Carl-Gustaf RossbyResearch Medalist. He was also recognized by the UnitedNations Environmental Program for contributing to the 2007Nobel Peace Prize to the IPCC.

Charles G. Bardeen is a research scientist in the AtmosphericChemistry Observations and Modeling laboratory at theNational Center for Atmospheric Research. His research inter-ests include the role of clouds and aerosols in climate and the

effects of energetic particles. In addition to studying polarmesospheric and cirrus clouds, he is also interested in abruptclimate change caused by extreme events including nuclearwar, meteor impacts, volcanic eruptions, and solar storms.

Lili Xia is a Research Associate in the Department ofEnvironmental Sciences at Rutgers University. Her researchinterests are climate change impacts on natural vegetation,agriculture, and air pollution.

Hans M. Kristensen is Director of the Nuclear InformationProject at the Federation of American Scientists inWashington, D.C., where he is responsible for researching anddocumenting the status and operations of nuclear forces of thenine nuclear-armed states. He is a frequent advisor to the newsmedia on the status of nuclear forces and policy. Kristensen isco-author of the bi-monthly FAS Nuclear Notebook column inthe Bulletin of the Atomic Scientists and the World NuclearForces overview in the SIPRI Yearbook, both of which aresome the most widely used reference material on the statusof the world’s nuclear arsenals.

Matthew McKinzie is Director of the Nuclear, Climate & CleanEnergy Program at the Natural Resources Defense Council(NRDC). He focuses on nuclear policy, specifically the conse-quences of reactor accidents and the ramifications of nuclearproliferation and nuclear war. At NRDC Matthew has alsoworked on other environmental issues such as climate change,renewable energy and national security, the harms of oil, gasand coal extraction, and clean water. Matthew holds a Ph.D. inexperimental nuclear physics from the University ofPennsylvania, and is currently an associate editor at the journalScience & Global Security. He is based in Washington, D.C.

R. J. Peterson is a Professor Emeritus of physics at theUniversity of Colorado and a former Jefferson Science Fellowfor the U.S. Department of State. After receiving his under-graduate (1961) and graduate (1966) degrees in Physics atthe University of Washington, he was an instructor atPrinceton University and on the research faculty at YaleUniversity. His research interests have covered many arenasof nuclear physics, including nuclear astrophysics, nuclear reac-tions, nuclear fission, and applications of nuclear reactions tocomputer memory elements. He has been a visiting professorat the University of Copenhagen (Niels Bohr Institute), theUniversity of Tokyo, and the Federal University of Rio deJaneiro. He is a Foreign Fellow of the Pakistan Academy ofSciences and a Fellow of the American Physical Society.

Cheryl S. Harrison is a biophysical oceanographer andAssistant Professor in Earth, Environmental and MarineScience at the University of Texas Rio Grande Valley, a 2018Make Our Planet Great Again fellow, and Earth system modelcoordinator for the Fisheries Model Intercomparison Project(FishMIP), part of the Inter-Sectoral Impact ModelIntercomparison Project (ISIMIP). She has had positions at theUniversity of Colorado Boulder, the University of CaliforniaSanta Barbara, the National Center for Atmospheric Research,and Oregon State University. Her research interests includephysical and biogeochemical ocean modeling, applied mathe-matics, marine ecology, fisheries, climate change, andgeoengineering.

Nicole S. Lovenduski is an oceanographer and AssociateProfessor at the University of Colorado Boulder. She studies

278 A. ROBOCK ET AL.

the biogeochemistry of the ocean with a particular focus onthe ocean carbon cycle and its role in the global climatesystem.

Richard P. Turco is a Distinguished Emeritus Professor ofAtmospheric and Oceanic Sciences at the University ofCalifornia, Los Angeles, and is the Founding Director ofUCLA’s Institute of the Environment and Sustainability. Turcoearned his Ph.D. in Electrical Engineering and Physics at theUniversity of Illinois, Urbana-Champaign, after receiving hisBachelor’s degree in Electrical Engineering at RutgersUniversity. His research has focused on the chemical andmicrophysical processes that control stratospheric ozone, theglobal climate system, and regional air pollution. Turco coinedthe term “nuclear winter” while leading the first collaborativeresearch team that defined the phenomena. His honors includea MacArthur Foundation Fellowship, and the Leo Szilard Awardfor Physics in the Public Interest of the American PhysicalSociety.

References

Coupe, J., C. G. Bardeen, A. Robock, and O. B. Toon. 2019. “NuclearWinter Responses to Global Nuclear War in the Whole

Atmosphere Community Climate Model Version 4 and theGoddard Institute for Space Studies ModelE.” Journal ofGeophysical Research: Atmospheres 124: 8522–8543.doi:10.1029/2019JD030509.

Mills, M. J., O. B. Toon, J. Lee-Taylor, and A. Robock. 2014. “Multi-decadal Global Cooling andUnprecedentedOzone Loss follow-ing a Regional Nuclear Conflict.” Earth’s Future 2: 161–176.doi:10.1002/2013EF000205.

Özdoğan, M., A. Robock, and C. Kucharik. 2013. “Impacts ofa Nuclear War in South Asia on Soybean and MaizeProduction in the Midwest United States.” Climatic Change116: 373–387. doi:10.1007/s10584-012-0518-1.

Toon, O. B., C. G. Bardeen, A. Robock, L. Xia, H. Kristensen,R. J. Matthew McKinzie, C. H. Peterson, N. S. Lovenduski, andR. P. Turco. 2019. “Rapid Expansion of Nuclear Arsenals byPakistan and India Portends Regional and GlobalCatastrophe.” Science Advances 5: eaay5478. doi:10.1126/sciadv.aay5478.

Xia, L., and A. Robock. 2013. “Impacts of a Nuclear War in SouthAsia on Rice Production in Mainland China.” Climatic Change116: 357–372. doi:10.1007/s10584-012-0475-8.

Xia, L., A. Robock,M.Mills, A. Stenke, and I. Helfand. 2015. “DecadalReduction of Chinese Agriculture after a Regional NuclearWar.”Earth’s Future 3: 37–48. doi:10.1002/2014EF000283.

BULLETIN OF THE ATOMIC SCIENTISTS 279